feat: complete T-530 cmhub URL normalization

app/appconfig.py

@@ -9,6 +9,7 @@ import copy
 import json
 import os
 import urllib.error
+import urllib.parse
 import urllib.request
 
 
@@ -129,6 +130,17 @@ def _deep_merge(defaults, loaded):
     return merged
 
 
+def _normalize_config_values(config):
+    if not isinstance(config, dict):
+        return config
+    ai = config.get("ai")
+    if isinstance(ai, dict):
+        cmhub = ai.get("cmhub")
+        if isinstance(cmhub, dict):
+            cmhub["base_url"] = normalize_cmhub_base_url(cmhub.get("base_url", ""))
+    return config
+
+
 def _assert_no_secrets(config):
     def visit(value, path):
         if isinstance(value, dict):
@@ -247,6 +259,7 @@ def save_config(config, path=CONFIG_PATH) -> dict:
     """Persist config to JSON and return the normalized config."""
 
     normalized = _deep_merge(DEFAULT_CONFIG, config)
+    _normalize_config_values(normalized)
     _assert_no_secrets(normalized)
     directory = os.path.dirname(os.path.abspath(path))
     if directory:
@@ -268,6 +281,7 @@ def load_config(path=CONFIG_PATH) -> dict:
         except json.JSONDecodeError as exc:
             raise ConfigError(f"配置文件不是有效 JSON: {path}") from exc
     normalized = _deep_merge(DEFAULT_CONFIG, loaded)
+    _normalize_config_values(normalized)
     _assert_no_secrets(normalized)
     return normalized
 
@@ -334,7 +348,7 @@ def cmhub_config(config=None) -> dict:
     if not isinstance(value, dict):
         raise ConfigError("ai.cmhub 必须是对象")
     merged = _deep_merge(DEFAULT_CONFIG["ai"]["cmhub"], value)
-    merged["base_url"] = str(merged.get("base_url", "") or "").strip()
+    merged["base_url"] = normalize_cmhub_base_url(merged.get("base_url", ""))
     merged["title_alias"] = str(merged.get("title_alias", "") or "").strip()
     merged["image_alias"] = str(merged.get("image_alias", "") or "").strip()
     merged["connect_timeout"] = int(merged.get("connect_timeout", 10) or 10)
@@ -344,8 +358,25 @@ def cmhub_config(config=None) -> dict:
     return merged
 
 
+def normalize_cmhub_base_url(base_url) -> str:
+    """Return the cmhub gateway root URL without path/query/fragment."""
+
+    text = str(base_url or "").strip()
+    if not text:
+        return ""
+    parts = urllib.parse.urlsplit(text)
+    if parts.scheme and parts.netloc:
+        return urllib.parse.urlunsplit((parts.scheme, parts.netloc, "", "", ""))
+    if parts.netloc:
+        return urllib.parse.urlunsplit((parts.scheme, parts.netloc, "", "", ""))
+    without_query = text.split("?", 1)[0].split("#", 1)[0].strip().strip("/")
+    if "://" not in without_query:
+        return without_query.split("/", 1)[0]
+    return text.rstrip("/")
+
+
 def cmhub_request_url(base_url, endpoint) -> str:
-    base = str(base_url or "").strip().rstrip("/")
+    base = normalize_cmhub_base_url(base_url)
     path = "/" + str(endpoint or "").strip().lstrip("/")
     if not base:
         return path